This study will test a theory-based intervention to increase levels of exercise among breast cancer survivors, in order to help them self-manage their long-term treatment side effects. The study focuses on outcome expectations (OEs), a significant yet understudied construct that is central to most behavior change theories. OEs refer to what people expect to obtain or avoid by engaging in a behavior; thus, people exercise because they believe it will produce desired outcomes. This randomized controlled trial will test a multicomponent intervention to increase exercise OEs and assess effects of OE changes on exercise intentions and levels of exercise. Eighty-eight early stage (IA to IIB) inactive breast cancer survivors will be recruited from the Duke Cancer Institute for a home-based exercise intervention. Participants will be randomized to the exercise OE intervention or an attention control group. Both groups will receive the American Cancer Society diet and exercise guidelines and an accelerometer. The intervention group will also receive an OE workbook containing self-directed activities to increase OEs by focusing on OE importance (value placed on the outcome(s)), certainty (perceived probability outcome(s) will occur); and accessibility (frequency with which the outcome(s) are thought of). The attention control group will receive a workbook containing the same activities focused on diet, rather than exercise. OE importance, OE certainty, OE accessibility, exercise intentions, exercise, and self-efficacy (a potential co-variate) will be assessed through online surveys at baseline, 4-, 8-, and 12-weeks post-intervention. At each time point, accelerometer data will be collected as an objective exercise measure. Specific aims of the study are to: (1) examine trajectories of OE importance, certainty and accessibility between participants in the OE intervention arm compared to participants in the attention control arm, across four time points: baseline, 4-, 8- and 12-weeks post-intervention; and (2) examine trajectories of exercise intentions and total weekly minutes of exercise, between participants in the OE intervention arm compared to participants in the attention control arm, across four time points: baseline, 4, 8 and 12 weeks post-intervention. If the intervention increases exercise among breast cancer survivors it may lead to improved survivorship duration and quality of life. The intervention can be implemented readily in clinic settings to support survivors to exercise more and thus experience better long-term outcomes. This study also provides a foundation for a developing a nurse scientist's program of research focused on designing theory-based exercise interventions for chronically ill populations.